Hybrid vigor in Eucalyptus increases resistance against Phytophthora root rot

Eucalyptus nitens is a cold-tolerant eucalypt that is native to Eastern Australia. Pure E. nitens as well as its hybrids, such as Eucalyptus grandis × Eucalyptus nitens, is propagated commercially in various regions of the southern hemisphere, including South Africa. In a plantation environment, E. nitens is susceptible to a variety of native and invasive pathogens, including Phytophthora alticola and P. cinnamomi. Recently, there have been increasing reports of root and collar rot in E. nitens in South Africa. The severity of this disease was substantially lower among interspecific hybrids of E. grandis × E. nitens compared to purebred E. nitens. In South Africa, the susceptibility of commercially propagated provenances of pure E. nitens and varieties of hybrid E. grandis × E. nitens to Phytophthora species is unknown. Therefore, we conducted greenhouse trials to evaluate the pathogenicity of P. alticola and P. cinnamomi to two families of pure E. nitens, one self-fertilized and the other outcrossed, as well as a single clonal variety of the most widely planted interspecific hybrid, E. grandis × E. nitens. The outcomes from these trials revealed that both self-fertilized and outcrossed families of E. nitens were highly susceptible to the tested Phytophthora species. The severity of root rot was greatest among plants inoculated with P. cinnamomi. The tested interspecific hybrid was tolerant to both Phytophthora species and developed new lateral and fine roots to offset the effects of root rot.


Introduction
Eucalyptus nitens is native to Eastern Australia. This tree is frequently planted in temperate regions of the southern hemisphere, such as in parts of Australia, Chile, New Zealand, and South Africa (Parsons et al. 2006;INFOR 2004;Lausberg et al. 1995;Purnell 1988). Eucalyptus nitens is often preferred over other cold-tolerant eucalypts because of its rapid growth (Clarke 2000;Pérez et al. 2006) and its frost hardiness (Pérez et al. 2006;Nakhooda and Jain 2016). This tree species is vulnerable to a wide range of pests and diseases (Roux et al. 2006;Carnegie et al. 2005;Adam et al. 2013), including various Phytophthora species, especially Phytophthora cinnamomi (de Andrade Lourenço et al. 2020;Studholme et al. 2019;Dick et al. 2006;Cahill et al. 2008).
Phytophthora species are important pathogens of trees in both natural and planted forests worldwide (Nagel et al. 2013;Burgess et al. 2017;Grünwald et al. 2012;Hansen 2008). Phytophthora cinnamomi, for example, affects the productivity of Eucalyptus plantations globally (Sena et al. 2018;Nagel et al. 2013;Burgess et al. 2021) while also causing serious damage to trees in native Eucalyptus woodlands (Dell and Malajczuk 1989;McDougall et al. 2002). Phytophthora species, including P. cinnamomi, P. alticola, and P. frigida, infect various Eucalyptus species in South Africa (Linde et al. 1994b;Maseko et al. 2007;Nagel et al. 2013;Bose et al. 2017). Due to the susceptibility of Eucalyptus fastigata and E. fraxinoides to P. cinnamomi, commercial deployment of these cold-tolerant species was substantially reduced in South Africa (Linde et al. 1994a;Wingfield and Kemp 1994).
One of the principal cold-tolerant tree species commercially propagated in South Africa is Eucalyptus nitens (Jones et al. 2004). Between 2010 and 2020, several incidences of root-and collar-rot-related mortality of E. nitens were observed. Several surveys and long-term research projects were launched to determine the factor(s) causing this previously unknown post-planting mortality among provenances of pure E. nitens (Jones 2019). Data from these studies indicated that the mortality rate of E. nitens was highest in the first year after planting, and it was significantly higher in lower altitude sites (Jones 2019). Both P. cinnamomi and P. alticola were isolated from the roots of some symptomatic trees (Plant Diagnostic Clinic, Forestry and Agricultural Biotechnology Institute). This disease outbreak, however, did not affect trees of the most commonly planted interspecific hybrid E. nitens × E. grandis (GN).
Field and greenhouse studies were conducted to evaluate the role of P. alticola and P. cinnamomi in the mortality of purebred E. nitens. In our greenhouse trials, a GN hybrid was included as a negative control because this variety was rarely affected by Phytophthora root and collar rot. We selected these two Phytophthora species because both were isolated from pure E. nitens trees affected by post-planting mortality. Based on field observations, we hypothesized that (1) the progeny from self-fertilized E. nitens would be more susceptible to P. alticola and P. cinnamomi than the outcrossed provenance, and (2) the interspecific hybrid would show tolerance to the tested Phytophthora species. In addition, based on published data (Bose et al. 2019;Maseko 2010), we hypothesized that (3) the severity of root rot would be highest among the plants inoculated with P. cinnamomi.

Monitoring of E. nitens plantations
Two E. nitens compartments (F15 and D38A) in Lothair, Mpumalanga Province (26° 24′ 18″ S, 30° 26′ 29″ E) were monitored to better comprehend the causes of the post-planting mortality. These two compartments were specifically chosen because the mortality rate was more than 30%. The seed lot numbers of the trees were used to identify their parents. Following that, microsatellite-based DNA fingerprinting was used to reconstruct the pedigree of 20 randomly selected trees, including 10 healthy and 10 symptomatic. The DNA fingerprinting of the sampled trees was outsourced to the Precision Tree Breeding Platform, Forest Molecular Genetics Programme, University of Pretoria (https:// www. fabin et. up. ac. za/ index. php/ resea rch-groups/ forest-molec ular-genet ics).

Greenhouse pathogenicity trials
Greenhouse trials were conducted to investigate the relative susceptibility of two different E. nitens families and an interspecific GN hybrid to two Phytophthora species. The pure E. nitens were from a self-fertilized family (selection 104; family 37656) and an outcrossed family (selection 131; family 37254). These two families were selected based on the outcomes of the field study. Seeds for both these pure E. nitens families were collected from Sappi's Eucalyptus seed orchard located in Lebanon, KwaZulu-Natal Province. DNA fingerprinting was performed to validate their genetic background. The interspecific hybrid (PP2107) was developed by the Institute for Commercial Forestry Research (ICFR), Pietermaritzburg, KwaZulu-Natal Province, and is the most widely planted GN hybrid in South Africa.

Selection of Phytophthora isolates
Isolates of P. alticola (CMW48711) and P. cinnamomi (CMW48774) were retrieved from the microbial culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria. Both isolates were previously used in a pathogenicity study of various Eucalyptus species (Bose et al. 2019) and originated from soil samples collected from a Eucalyptus grandis plantation located in Commondale, Mpumalanga Province (S27° 17′ 39″ E30° 53′ 56″).

Assessment of virulence
After retrieving both Phytophthora isolates from the culture collection, the virulence of each species was reassessed with an under-bark inoculation method using E. nitens as the host. Twenty-one days after inoculation the pathogens were re-isolated from infected plant tissue onto a Phytophthora-selective medium, NARPH (50 mg nystatin, 200 mg ampicillin, 10 mg rifampicin, 25 mg pentachloronitrobenzene, and 50 mg hymexazol per 1 L of deionized water and 15 g cornmeal agar) as described by Hüberli et al. (2000). The identity of the isolated pathogens was determined by amplifying the complete internal transcribed spacer region using primers ITS6 and ITS4 (Cooke et al. 2000;White et al. 1990). These reisolated cultures were used for subsequent inoculation studies.

Seed germination and transplanting of seedlings
Seeds of the self-fertilized and outcrossed families of E. nitens were germinated under natural light in steam-sterilized polystyrene trays with autoclaved composted pine bark medium. Thereafter, the trays were incubated in covered tunnels (mean temperature 25 °C; mean relative humidity 65%) at the Sappi Shaw Research Centre, Howick, KwaZulu-Natal Province. All trays were irrigated once every day. The plants were fertilized with Osmocote® controlled release fertilizer pellets that were sprinkled on the surface of the potting medium.
Plants of the interspecific GN hybrid were propagated from shoot cuttings. Selected shoot segments were allowed to root in a sterile mixture of coir and perlite (9:1) in an enclosed rooting tunnel where the relative humidity was set between 60 and 70% and the mean temperature ranged between 18 and 24 °C with underfloor heating. All the cuttings were irrigated with overhead misting at regular intervals throughout the day. Before initiation of the trial, the rooted cuttings were hardened in a separate covered tunnel for a period of 5 to 6 weeks under ambient temperature.
For each pure family, one E. nitens seedling measuring 5-8 cm in height was transplanted into sixty 0.5 L pots containing sterile sand. For the interspecific GN hybrid, a single-rooted cutting was transferred into a pot with sterile sand. Two plastic tubes measuring 10 × 1.5 cm were inserted into each pot such that each can receive approximately 2.5 g of Phytophthora inoculum.
Irrespective of family, all pots had one seedling of approximately the same height. All the plants were allowed to grow in a covered tunnel with temperatures ranging between 23 and 28 °C and relative humidity of 50-60%.

Preparation of inoculum
For both Phytophthora species, the inocula were prepared in vermiculite as described by Bose et al. (2019). Each flask contained 600 mL of sterile growth medium (600 mL coarse vermiculite from Culterra, South Africa; 10 g millet seeds; 170 mL of 10% V8 juice, Campbell Soup Company, USA; 150 mL of deionized water; and 2 g calcium carbonate, Sigma-Aldrich, USA). The flasks were inoculated with Phytophthora cultures grown on 10% V8-agar (100 mL clarified V8 juice; 900 mL deionized water; 15 g Difco Agar, Becton, Dickinson and Company, Sparks, USA) for 8 days in darkness. The inoculated Erlenmeyer flasks were incubated in darkness at 20 °C for 6 weeks. During this time, the flasks were gently rotated at regular intervals for even distribution of the mycelia.

Inoculation of plants
Three months after transplanting into pots, seedlings of the E. nitens families and clonal variety of the GN hybrid were inoculated with the Phytophthora species. For this, both plastic tubes were removed from the pot, and approximately 2.5 g of the Phytophthora inoculum was dispensed into each cavity. For negative controls, the cavity was filled with an equal amount of sterile vermiculite medium. The cavities were then sealed with sterile sand.

Experimental design
The first replicate of the trial was conducted in 2019. Seedlings and cuttings were transplanted in June, inoculated in September, and harvested in December. In 2020, the second replicate of the trial was completed. The seedlings and cuttings were transplanted in April, inoculated in July, and harvested in October. Regardless of the trials, the first week of each month was dedicated to transplanting, inoculating, and harvesting.
In both repeats of the trial, there were 20 biological replicates per treatment (control, P. alticola, and P. cinnamomi) for each host genotype (self-fertilized, outcrossed, and interspecific hybrid). After inoculation, pots were arranged randomly in plastic trays on a bench in the covered tunnel. The negative controls were arranged in separate trays from trees that received Phytophthora inoculum. The arrangement of the pots within trays was changed regularly to reduce any microclimate effect. The plants were flooded three times: immediately after inoculation, then after 14and 28-day post-inoculation, by filling the trays with water. All the plants were irrigated once every day until they were harvested.

Measurement of symptoms
Seedlings and cuttings were carefully harvested 3 months after inoculation. This was done by rinsing off all the growth media from the roots, under running tap water. The severity of root rot was evaluated on a scale of 0-4 (0 = no visible root damage, 1 = ~ 20% roots with lesions, 2 = > 20% roots with lesions, 3 = > 50% roots with lesions, 4 = dead). During harvesting, the fresh shoot and root masses and root volumes were measured using a volume displacement method (Harrington et al. 1994). For the dry weights of shoots and roots, plants were dried at room temperature (21-24 °C) for a period of 30 days and then weighed using an analytical scale.

Re-isolation and identification of Phytophthora species
To fulfil Koch's Postulates, root sections with lesions from all plants were plated onto the Phytophthora selective medium NARPH. Mycelia emerging from plated root tissues were subcultured onto half-strength potato dextrose agar medium (19.5 g PDA powder, Merck, South Africa; 7 g Difco agar; 1 L of deionized water). Molecular identification of a selection of isolates was done by amplifying the complete internal transcribed spacer (ITS) region of the rDNA using the primer pair ITS6 and ITS4. The sequences were identified using the BLAST algorithm (Altschul et al. 1990) available through the NCBI GenBank.

Statistical analyses of datasets
All datasets were statistically analyzed using R v4.2.1 (R Core Team 2021). A Wilcoxon signed rank test with continuity correction was used to determine whether the genetic background of pure E. nitens affected post-planting mortality. Growth parameter data from the two replicate trials could not be transformed towards normality and was analyzed using the Kruskal-Wallis chi-squared test. Differences between treatments were consistent between the two trials. For analysis of the combined dataset, data from trial 2 was normalized towards the mean values of trial 1 using a factor of 5 for fresh shoot weight, dry shoot weight, and root volume, and a factor of 9 for fresh and dry root weight. The normalized dataset was analyzed using the Kruskal-Wallis chi-squared test, and differences between individual treatments were determined using Dunn's test. Disease severity data were analyzed using the Kruskal-Wallis chi-squared test, and differences between individual treatments were determined using model simplification.

Monitoring of E. nitens in plantations
Significant differences were found in the number of E. nitens trees affected by the post-planting root rot disease in the two compartments (F15 and D38A) monitored for the disease. The results of DNA fingerprinting revealed a significant relationship between the survival of E. nitens in the field and their genetic background. Post-planting mortality affected self-fertilized families more than outcrossed families (p = 0.001904; Fig. 1).

Greenhouse pathogenicity trials
DNA fingerprinting data of purebred E. nitens showed that selection 131 was completely outcrossed, whereas the selfpollination percentage in selection 104 ranged from 40 to 100%.
In both outcrossed and self-fertilized E. nitens seedlings, compared to the controls, both pathogens significantly reduced fresh and dry shoot weights, fresh root weights, and root volumes (Table 1; Fig. 3). However, the dry root weight of outcrossed and self-fertilized seedlings produced variable results with only one of the treatments resulting in significant differences for each (Table 1; Fig. 3). For the interspecific hybrid, only dry root weight and root volumes were significantly reduced compared to the controls. Although fresh and dry shoot weights and fresh root weights were reduced, the differences with the controls were not significant (Table 1; Fig. 3). While P. cinnamomi-infected GN clones had decreased dry root weight and volume, clones that were infected with P. alticola had significantly reduced dry shoot and root weights, and root volume (Table 1; Fig. 3). In all instances E. nitens plants were more severely affected by the pathogens than the GN hybrid.
When the parameters of pathogenicity were compared between the three genotypes, they were significantly different across all the treatments (Table S1). However, when these same parameters were compared between outcrossed and self-fertilized E. nitens, differences were negligible (Table S1). This showed that disease severity was nearly uniform between the pure families of E. nitens.  Fig. 1 Eucalyptus nitens survival after planting in the field is influenced by their genetic background. DNA-fingerprinting data revealed that self-fertilized E. nitens provenances were more susceptible to post-planting mortality compared to outcrossed provenances

Re-isolation and identification of Phytophthora isolates
In both trials, the greatest number of Phytophthora isolates were recovered from self E. nitens seedlings followed by the outcrossed family and then the GN hybrid (Fig. 4). Based on consolidated results from both trials, P. cinnamomi was most consistently recovered from the inoculated plants (Fig. 4).

Discussion
Pathogenicity trials were conducted to assess the susceptibility of a self-fertilized and outcrossed provenance of E. nitens and an interspecific GN hybrid to infection by P. alticola and P. cinnamomi. Statistical analyses of the data emerging from these pathogenicity trials showed that both self-fertilized and outcrossed E. nitens seedlings were equally susceptible to both pathogens. In contrast, the interspecific GN hybrid  showed higher levels of tolerance to infection by both of the tested Phytophthora species. Under plantation conditions, we observed that the selffertilized E. nitens family was more susceptible to postplanting mortality compared to the outcrossed selection. Self-fertilization is expected to increase homozygosity, which can result in inbreeding depression (Cheptou 2018;Pupin et al. 2019). Hence, it was not surprising that the self-fertilized E. nitens seedlings were more severely affected by post-planting mortality under field conditions. In contrast, outcrossing promotes the emergence of new traits. Consequently, tree breeders frequently utilize outcrossing for plant improvement (Chen et al. 2021). Yet, in our greenhouse inoculation trials, there were no statistically significant differences in susceptibility to inoculation with Phytophthora species between self and outcrossed seedlings. This could be due to a higher or more consistent inoculum load in the greenhouse trials compared to the plantation situation.
The interspecific GN hybrid displayed tolerance towards infection by P. alticola and P. cinnamomi. Among eucalypts, interspecific hybridization is promoted by their remarkably conserved genomic structure (Butler et al. 2017). These hybrids with novel gene combinations can include desirable traits from both parents, including rapid growth and disease resistance (Teixeira et al. 2009;Guimarães et al. 2010;Bradshaw and Grattapaglia 1994). It remains unknown whether interspecific hybridization among Eucalyptus species can confer resistance to Phytophthora diseases. However, in other trees, QTLmediated resistance to P. cinnamomi has been achieved by backcrossing American and Chinese chestnuts (Castanea species), where the latter was resistant to the pathogen (Zhebentyayeva et al. 2019). This might be partially true for GN because, in a previous trial, E. grandis showed tolerance towards P. alticola but not to P. cinnamomi (Bose et al. 2019;Maseko 2010).
Tolerance of the GN hybrid to infection by the two tested Phytophthora species in the study could also be due to unique phenotypic characteristics in the hybrid and not in the parent plants. For example, this tolerance could stem from the rapid growth traits inherited from E. grandis, allowing the hybrid variety to endure the impact of the root rot by producing new lateral and fine roots to offset the effect of the pathogens. A similar situation was seen in an earlier study, where P. multivora-infected E. grandis also displayed a defense response by producing a considerably larger root mass and root-to-shoot ratio than the control (Bose et al. 2019).
In this trial, the severity of root rot was greater in plants infected with P. cinnamomi than in those infected with P. alticola. Both of these Phytophthora species, however, are known to induce root rot in E. nitens (Maseko 2010). The observed differences might be attributed to the E. nitens provenances used in this study, which may have been less susceptible to P. alticola than to P. cinnamomi. The lower level of susceptibility might also account for the fact that we re-isolated P. alticola from the roots of the GN hybrid Table 1 Medians and statistical significance (p value) of quantitative measurements for disease severity in inoculated and control genotypes of Eucalyptus nitens seedlings and an interspecific hybrid (GN). The data was analyzed using the Kruskal-Wallis chi-squared test, and differences between different treatments were determined using Dunn's test. Letters next to the numbers denote statistical differences (n = 20) Page 7 of 10 24 that did not exhibit root rot symptoms. This could be attributed to the fact that some Phytophthora species can induce asymptomatic infections (Migliorini et al. 2019;Belhaj et al. 2018). Similarly, in our previous trials involving E. grandis and A. mearnsii, P. alticola was also re-isolated from the asymptomatic roots of both hosts (Bose et al. 2019). This suggests that isolating P. alticola from the roots of trees and soils does not necessarily equate to its role in tree decline. Phytophthora root rot can be a serious constraint to Eucalyptus propagation in South Africa. Results of this study have revealed opportunities to reduce this problem through the utilization of Eucalyptus hybrids such as those between E. grandis and E. nitens. It is also relevant that we observed Fig. 3 Comparison of the shoot and root weights and root volumes between provenances of self-fertilized and outcrossed E. nitens and an interspecific hybrid (E. grandis × E. nitens) inoculated with Phytophthora alticola or with P. cinnamomi. Controls were mock-inoculated with sterile vermiculite inoculum. Box plots were constructed using combined, normalized datasets from two trials. Asterisks in red indicate statistical significance (p ≤ 0.05; Kruskal-Wallis chi-squared test; n = 20) serious root rot among the pure families of E. nitens in our trials, but no mortality. This is likely due to our trials being conducted in greenhouses where the plants were regularly irrigated and fertilized. In the plantation situation, however, loss in root structure combined with drought substantially increases mortality. This would also be consistent with the fact that the most severe post-planting mortality of E. nitens occurred in 2016 when severe drought was experienced, similar to the decline of oak and beech in Italy (Colangelo et al. 2018;Seddaiu et al. 2020) and Central Europe (Jung 2009), respectively, as well as other parts of the Mediterranean region (Peñuelas and Sardans 2021) where Phytophthora is causing severe losses during extreme climatic events.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.